Applied AI-Powered Bioinformatics Tools

Build a Sequence Analysis Dashboard

Last reviewed

Beginner

What you'll learn

~40 min
  • Ask an AI tool, in plan mode, for a page that reads a FASTA file (a text file of named DNA sequences) and shows each sequence's length, GC content, base counts and open reading frames
  • Read its plan for the counting rules: lower case counted as capitals, ambiguity letters such as N counted on their own, open reading frames on both strands, file totals added from the bases
  • Check the page's numbers against values worked out for the practice file
  • See why a GC percentage can look ordinary for a sequence that is half unknown

The story, and what you will end up with

A lab your facility serves sends three DNA sequences and asks for a quick look before they decide on their next step: how long each one is, its GC content, and whether it holds an open reading frame. You could count letters in a spreadsheet. A small page that does it for any FASTA file is quicker, and it counts the same way every time.

By the end of this lesson you will have a sequence dashboard on your own computer: a page in your browser that reads a FASTA file and shows, for each sequence, its length, GC content, base counts and open reading frames, with totals for the whole file. It runs on your computer and sends nothing anywhere. You build it by asking your AI tool in plain English, and you check it against numbers this lesson gives you.

🔍Words used in this lessonclick to expand
  • FASTA: a text file of sequences. Each one starts with a header line beginning with >, followed by its letters, often wrapped over several lines. The files in this lesson are DNA. A file of RNA uses U where DNA has T; this page is not asked to handle RNA, though you could ask for that later.
  • GC content: the share of a sequence’s bases that are G or C, as a percentage. In this lesson it is counted over A, C, G and T only. That is a convention, the one this page uses. Other conventions exist: some programs also count S (G or C) as GC and W (A or T) as AT, since each is certain on that point. What matters is that the page prints which convention it used.
  • Ambiguity letters: letters that stand for more than one base. N means any base, so the base at that position is not known; R means A or G; Y means C or T.
  • Open reading frame (ORF): here, a stretch that starts at ATG and runs, three letters (one codon) at a time, to the first stop codon in the same frame: TAA, TAG or TGA.
  • Six frames: a sequence can be read from its first, second or third letter, on the strand as written and on its reverse complement, the opposite strand read the other way.

Before you start

First, if your AI tool is still open, leave it: type /exit and press Enter (if you use Antigravity, press Ctrl+D with nothing typed instead), and wait until you see the $ or % prompt. The next line goes there, in the terminal, not in the AI tool. It makes a folder for this lesson beside your practice folder and goes into it:

Terminal window
mkdir -p ~/bio-workbench/01-sequence-dashboard && cd ~/bio-workbench/01-sequence-dashboard

Start your AI tool with the start line you kept from the setup lesson, the one with approval prompts off. Then ask it to set the folder up:

Set this folder up so we can undo any change you make, download https://learn.intellimetrics.net/starters/biotech/01/samples.fasta into a data folder, and save a checkpoint.

Look for it saying three things: the folder is set up with Git (the program that saves snapshots, from setup), the file is saved as data/samples.fasta, and a first snapshot, your checkpoint, is saved. Later you can ask whether the file has changed since then. If it does not mention one of them, ask: “Did you save a checkpoint with the file in it?” If the download does not work, download samples.fasta in your browser, check where it went (the browser’s list of downloads shows the folder), and tell your AI tool, for example: “samples.fasta is in my Downloads folder. Copy that one file into a data folder here, and save a checkpoint.” In Windows PowerShell or a desktop app, “my Downloads folder” works as is; only say “my Windows Downloads folder” if you use Ubuntu (WSL) on Windows, which keeps its own Downloads folder apart from Windows.

The three sequences are random letters made up for this course, not from any organism or sample.

Open this box once, because it says what each sample name and term used further down stands for.

ℹWhat is in samples.fastaclick to expand

Three sequences, PRAC-01 to PRAC-03, each with a header line such as >PRAC-01 practice sequence, invented:

SequenceLengthHow it is writtenLetters other than A, C, G, T
PRAC-01420Capitals, wrapped at 60 letters a lineNone
PRAC-02360Wrapped at 70 letters a line; bases 1 to 60 and 301 to 360 in lower case; a blank line after itNone
PRAC-03240Capitals, wrapped at 60 letters a lineSeven: N at bases 15 to 18, R at 53 and 78, Y at 99

Lengths count every letter of the sequence, ambiguity letters included, and not the header line.

Switch to plan mode

Your AI tool is already open. Switch to plan mode now: Shift+Tab until it says plan, or /plan (the keys for each tool). In plan mode it usually reads your file, asks its questions and shows you a plan before it builds anything; the order can vary.

The conversation

Each message below is an example. Say it your own way.

Say what you want

Say what you do, which file to use, and what you want to see:

I check DNA sequences for labs at our core facility. Help me build a small browser dashboard showing each sequence's length, GC content, base counts and open reading frames, for files like data/samples.fasta. Ask me what you need first.

Your AI tool usually reads the file and comes back with questions, sometimes a dozen at once, each with a suggested answer. It may also remark on the lower-case letters in PRAC-02.

Answer its questions

Answer each in a sentence. These are questions the tools often ask, with one way to answer each. Yours may ask them in a different order or show a list of choices:

If it asksYou might say
Minimum length for an ORF, which start codons, which strands, which genetic code?At least 100 codons, ATG to the first stop, all six frames, standard genetic code.
Lower-case letters: ordinary bases, or shown apart?Count lower case as ordinary bases.
N and other ambiguity letters: their own count, left out of GC?Yes. Keep them in their own count, out of GC.
ORFs that reach the end of a sequence with no stop codon?List them apart, marked as having no stop, and don’t count them as ORFs.
How many sequences, and how long?A few short sequences at a time.
Where will it run? Shared with other staff?On my computer only.
How will you give it a file?Show data/samples.fasta at start, with a button to open another FASTA file.
Export to a spreadsheet file, or on screen?On screen is enough for now.

For these practice sequences lower case means nothing, so counting it as ordinary bases is right here. In a real FASTA file it may carry meaning, such as repeats masked by an earlier program or bases of low quality, and folding it into capitals throws that away; ask whoever produced the file, and have the page say how it treated lower case.

If it asks several questions at once, answer the ones about how things are counted in a sentence, and for the rest say that its suggestions are fine. If it asks something not listed here, answer it the same way, in a sentence.

Read the plan

Next it usually shows you a plan: what it will build and how it will check it. Read it for six things:

  1. Every sequence in the file gets its own results, including one wrapped over several lines and one after a blank line.
  2. Lower case counts the same as capitals.
  3. GC content is counted over A, C, G and T only, and N and the other ambiguity letters have a count of their own, shown beside it. The page says which convention it used.
  4. ORFs run from ATG to the first stop in the same frame, on both strands, at least 100 codons, and the plan says whether the stop codon is counted in the length.
  5. The totals for the whole file add up the bases, and the file’s GC content is worked out from those, not averaged from the three percentages. An average would weigh a short sequence the same as a long one.
  6. Your sequences stay on this computer, and your file is read and left as it is.

A first plan often leaves one of these out, most often the file totals in item 5. Say so in a sentence, in your own words, for example:

Also show totals for the whole file: all the bases added together, and a GC content worked out from those bases, not an average of the percentages.

When the plan has all six, approve it: if your AI tool offers a choice to go ahead, pick it; otherwise say “Yes, go ahead and build it.” In Codex, press Shift+Tab first to leave plan mode. In Copilot, saying it brings up a choice to go ahead; pick that. Let it build. It may test what it built and tell you how to open it. When it has finished, ask it: “Did you install anything, and where?” Nothing, or something inside this lesson’s folder, is fine for these lessons. If it names anywhere else on your computer, such as a few scratch files of its own, ask it what they are for, and to remove them if you do not want them there.

Look at what it built

Ask it to open the page:

Open the dashboard in my browser.

A page opened straight from a file cannot fetch other files by itself, such as the practice file it shows at start, so your AI tool will probably start a small program on this computer that serves the page at a local address, one ending in a number, such as http://127.0.0.1:8765/ (127.0.0.1 is an address that means this computer). If it gives you an address without opening it, copy the address into your browser’s address bar and press Enter. If it says the address is already in use, say: “Something else is using that address. Don’t stop it. Use a different address for my page.” Leave the program running while you look; when you are done, ask your AI tool to stop it. If your AI tool does not take a message after opening the page, wait a minute. If it still does not, close the terminal window, open a new one, type this lesson’s folder line again, start your tool, and say: “Start the page again so it keeps running while we talk.” The page and your saved work are files in the folder, so nothing is lost. If the page says it cannot load the file, tell your AI tool what the page says. Your page will be laid out in its own way.

Check it for these:

  • There are three sequences, PRAC-01, PRAC-02 and PRAC-03, with lengths 420, 360 and 240.
  • GC content is 46.90% for PRAC-01, 51.94% for PRAC-02 and 51.07% for PRAC-03 (46.9, 51.9 and 51.1 if your page shows one decimal place). If PRAC-02 shows a different figure, check first whether lower-case letters are being counted as bases.
  • PRAC-03 shows seven ambiguity letters: four N, two R and one Y. Its GC content is 119 G or C out of 233 A, C, G or T. If it shows 49.58%, the ambiguity letters have gone into the count they should be left out of.
  • The whole file has 1,020 bases, and its GC content is 49.65%: 503 G or C out of 1,013 A, C, G or T. If it shows 49.97%, the three percentages have been averaged.
  • PRAC-01 has one ORF, on the strand as written, starting at base 61: 110 codons not counting the stop codon, 111 counting it. It ends at base 393, the last letter of its stop codon.
  • PRAC-02 has one ORF, on the reverse strand: 104 codons not counting the stop codon, 105 counting it, 315 bases from the A of ATG to the last letter of the stop. Read along the reverse strand, it begins ATGCGAGGA and ends CTGAAATAA. On the sequence as written it covers bases 21 to 335. Pages number the reverse strand in different ways, so yours may show 335 to 21, or 26 to 340 counted along the reverse strand. Whatever the numbering, the strand, the length in bases and in codons, and those first and last letters must agree.
  • PRAC-03 has no ORF of 100 codons or more.
  • PRAC-03’s reading frames run across N, R and Y. A codon with an ambiguity letter may not be translatable with certainty, so the page should say how it treated one: as unknown, as ending the frame, or some other way. If it does not say, ask: “What does the page do with a codon that has N, R or Y in it?”
  • When you ask “Has my sequence file changed since the checkpoint?”, it compares the file with the checkpoint and says no.

If something on the list is not true, tell your AI tool in a sentence what you see and what you expected, for example: “PRAC-03 shows 49.58% GC. Leave N, R and Y out of the GC count, and show them in their own column.”

Notice what the percentage leaves out

Look at PRAC-03’s GC content beside the other two. Nothing about it stands out. Now try this:

Make a copy of data/samples.fasta in this folder. In the copy only, change the last 120 bases of PRAC-03 to N. Then show me PRAC-03's length and GC content from the copy beside the original.

Before you read on: half of PRAC-03 is now unknown. What do you expect its GC content to do?

It barely moves. In the copy, PRAC-03 is still 240 bases long, and its GC content goes from 51.07% to 51.33%. The length counts every letter, N included. The GC content is now worked out from the 113 bases that are still A, C, G or T, about half as many as before, and those happen to have much the same share of G and C. The file’s total GC content moves only from 49.65% to 49.50%. The one number that changes sharply is the count of ambiguity letters: from 7 to 127.

So a sequence that is half unknown can show an ordinary length and an ordinary GC content. The count of N and other ambiguity letters is the column that tells you, and only if you look at it.

This page counts a GC percentage over A, C, G and T only, so read it beside the count of N and other ambiguity letters before you trust it.

Whether a sequence like that is fit to report back to the lab is your call, not the page’s. When you have looked, remove the copy: “Delete that copy; I’m done with it.”

One real run

Here is part of what one AI tool said after the experiment above, with Claude Code on 11 November 2026. Yours will word it differently; check your page against the list above, not against this.

- Length: stays at 240, because N still counts towards length.
- GC content: barely moves, because N is left out of the calculation.
It's now based on only the 113 bases that are still A, C, G or T,
about half of what it was.

The sequences it talks about, and their names PRAC-01 to PRAC-03, are made up for this course.

How real sequences differ

The practice sequences are random letters, so nothing about them is biology: their GC content and their reading frames mean nothing beyond this exercise.

  • Real files can carry letters this file does not: other ambiguity letters, gaps, or characters a page does not expect. Ask your AI tool what the page does with a letter it does not recognise, and have it show that letter in a count of its own rather than drop it.
  • Lower case may mean something in a real file, such as masked repeats or bases of low quality. Ask whoever produced the file, and check that the page says how it treated lower case.
  • Programs differ in how they count. Whether a length includes N, whether S and W count towards GC, whether an ORF’s length includes the stop codon, how the reverse strand is numbered: two programs can disagree on the same file and both be right by their own rule. When a number goes back to a lab, say which rule it was counted by.
  • Real sequences belong to the labs that sent them, and some come from human samples. They do not go into this folder or to your AI tool; the steps under Keep going show how to use the finished page on them.

Keep going

Small changes do not need plan mode. Say each in a sentence, then look at the page again:

  1. “Mark where EcoRI (GAATTC) and BamHI (GGATCC) sites fall in each sequence.” On the practice file there is one EcoRI site, in PRAC-03, starting at base 191, and no BamHI site.
  2. “Show the codon usage for the longest ORF in each sequence.”
  3. “Show each sequence’s share of ambiguity letters as a percentage beside its GC content.”
  4. “Save a checkpoint.” Your AI tool saves a snapshot you can return to.

When you want to look at a real sequence file, use the finished page. It runs in your browser by itself, with your AI tool closed, so your AI tool never sees the file. Do these steps in order; the first three happen now, while your AI tool is open and this folder still holds only the practice file:

  1. Go through the data check, the questions in setup about what the data is, who decides how it is used, and which AI tool would see it. A lab’s sequences are that lab’s to decide on, and human-sample data may need your facility director or compliance office as well. Go on only if they allow this use.

  2. Decide on a folder of its own for your real sequence files, outside this course folder, for example ~/lab-sequences. Nothing has to be in it yet. Ask your AI tool for the exact line that starts the page and the address it opens at, for example: “Write me the exact line to start this page myself, and its address, so I can open a FASTA file from ~/lab-sequences.” Use your own folder’s name, written out in full from your home folder as above. If it offers to copy your real files into this folder instead, say no.

  3. Copy the line and the address it gives you into a note. They will look something like these, though yours may differ:

    Terminal window
    python3 ~/bio-workbench/01-sequence-dashboard/serve.py

    and the address http://127.0.0.1:8765/. Read the line left to right: python3 starts Python, which runs the program; next comes the program your AI tool wrote to show the page, in this lesson’s folder. ~ means your home folder; your line may write it out in full instead. The address is where the page appears in your browser while that program runs. If your page opens straight from its file, you may get the file’s location instead of a line; write that down. Write down too which button on the page opens a file, such as Open FASTA. If you are unsure about any part, ask your AI tool now, while it is still open.

  4. Ask your AI tool to stop anything it started for the page, then leave it the way you did at the start of this lesson.

  5. Put your real files in that folder. At the $ or % prompt, type this lesson’s folder line again, then paste the line from your note and press Enter. Open the address in your browser and choose your real file with the page’s button; the page then reads it on your computer and saves nothing.

Try another case

Before the quiz, write two or three sentences for yourself. A lab asks why your dashboard says their sequence is 1,200 bases long with 41% GC, while the program they use says 1,180 bases and 40% GC. What would you check first, and what would you tell them?

A good answer looks for letters the two programs count differently before suspecting either one: ambiguity letters, lower case, gaps or stray characters in the sequence. It asks what each program counts in its length and in its GC content, and it tells the lab which rule your number was counted by.

Key takeaways

  • Ask briefly, and let your AI tool ask back. Say what you do, which file to use and what you want to see; its questions fill in the counting rules.
  • Read the plan for the counting rules, not the layout: lower case, ambiguity letters, both strands, totals added from the bases.
  • Check a new page against numbers you know before you trust it on a file you do not.
  • A GC percentage can look ordinary for a sequence that is half unknown. The count of ambiguity letters is what shows it.
  • Practice data stays practice data: a lab’s real sequences go to the finished page, with your AI tool closed, after the data check.
KNOWLEDGE CHECK

Your dashboard shows a sequence 600 bases long with 50.00% GC content. Its count of ambiguity letters is 400, all N. What does the 50.00% describe?

What’s next

The next lesson is about checking a genome-editing core’s exported candidate records against the order they were made for, and listing what needs a look.

intellimetrics · Arlington, VA
© 2026 intellimetrics.net
Search lessons